bdgraph.sim: Synthetic graph data generator

Description

This function implements a synthetic graph data generation for multivariate distributions with different types of underlying graph structures, including "random", "cluster", "scale-free", "hub", "fixed", and "circle". Based on the underling graph structure, it generates four different types of datasets, including multivariate Gaussian, non-Gaussian, discrete, or mixed data.

Usage

bdgraph.sim( n = 2, p = 10, graph = "random", size = NULL, prob = 0.2, 
             class = NULL, type = "Gaussian", cut = 4, b = 3, D = diag(p), 
             K = NULL, sigma = NULL, mean = 0, vis = FALSE )

Arguments

The number of samples required. The default value is 2.

The number of variables (nodes). The default value is 10.

graph

The graph structure with option "random" (as a default), "cluster", "scale-free", "hub", "fixed", and "circle". It also could be adjacency matrix corresponding to a graph

size

The number of links in the true graph (graph size).

prob

For a "random" graph, it is the probability that a pair of nodes has a link. The default value is 0.2.

class

Number of classes for option "cluster".

type

Type of data with four options "Gaussian" (as a default), "non-Gaussian", "discrete", and "mixed". For option "Gaussian", data are generated from multivariate normal distribution. For

cut

Number of categories for simulating discrete data (type = "discrete"). The default value is 4.

The degree of freedom for G-Wishart distribution, $W_G(b, D)$. The default is 3.

The positive definite $(p \times p)$ "scale" matrix for G-Wishart distribution, $W_G(b, D)$. The default is an identity matrix.

A positive-definite symmetric matrix specifying the precision matrix. It is for option graph = "fixed".

sigma

A positive-definite symmetric matrix specifying the covariance matrix. It is for option graph = "fixed".

mean

A vector specifying the mean of the variables. The default value is a zero vector.

vis

Visualize the true graph pattern. The default value is FALSE.

Value

An object with S3 class "simulate" is returned:
dataGenerated data as a $n \times p$ matrix.
sigmaThe covariance matrix of the generated data.
KThe precision matrix of the generated data.
GThe adjacency matrix corresponding to the true graph structure.

References

Mohammadi, A. and E. Wit (2015). Bayesian Structure Learning in Sparse Gaussian Graphical Models, Bayesian Analysis, 10(1):109-138 Mohammadi, A. and E. Wit (2015). BDgraph: An R Package for Bayesian Structure Learning in Graphical Models, Arxiv preprint arXiv:1501.05108v2 Mohammadi, A., F. Abegaz Yazew, E. van den Heuvel, and E. Wit (2015). Bayesian Modeling of Dupuytren Disease Using Gaussian Copula Graphical Models, Arxiv preprint arXiv:1501.04849v2

Examples

Run this code

# generating multivariate normal data from a 'random' graph
data.sim <- bdgraph.sim( n = 50, p = 10, prob = 0.3, vis = TRUE )
print(data.sim)

# generating multivariate normal data from a 'hub' graph
data.sim <- bdgraph.sim( n = 3, p = 6, graph = "hub", vis = FALSE )
data.sim $ data

# generating mixed data from a 'hub' graph 
data.sim <- bdgraph.sim( n = 10, p = 8, graph = "hub", type = "mixed" )
data.sim $ data